CEVA Inc (CEVA) 2018 Q2 法說會逐字稿

Good morning, and welcome to the CEVA Incorporated Second Quarter 2018 Earnings Conference Call. (Operator Instructions) Please note this event is being recorded.

I would now like to turn the conference over to Richard Kingston, VP of Market Intelligence, Investor and Public Relations.

Thanks, Brandon. Good morning, everyone, and welcome to CEVA's Second Quarter 2018 Earnings Conference Call. I'm joined today by Gideon Wertheizer, Chief Executive Officer of CEVA; and Yaniv Arieli, Chief Financial Officer of CEVA. Gideon will cover the business aspects and the highlights from the second quarter and provide general qualitative data. Yaniv will then cover the financial results for the second quarter and will also provide qualitative data for 2018.

I will start with the forward-looking statements. Please note that today's discussion contains forward-looking statements that involve risks and uncertainties as well as assumptions that if they materialize or prove incorrect could cause the results of CEVA to differ materially from those expressed or implied by such forward-looking statements and assumptions. Forward-looking statements include our revised financial guidance for full-year 2018 revenues, third quarter and annual royalty revenues, optimism about a healthy licensing environment and demand for CEVA's products, optimism about the distinguishing features of CEVA's AI edge technology, positive forecasts from Tractica forecast and Bluetooth SIG forecast and optimism about CEVA's technology is being adopted by ZTE at a later time.

For information on the factors that could cause a difference in our results, please refer to our filings with the Securities and Exchange Commission. These include the ability of CEVA's signal processing IPs for smarter connected devices to continue to be strong growth drivers for us; the traction with edge technology for AI; our success in penetrating new markets and maintaining our market position in existing markets; the ability of new products incorporating our technologies to achieve market acceptance and offset the maturity of the handset markets; the speed and extent of the expansion of the LTE and 5G networks, AI, LTE IoT and IoT space generally; our ability to execute more broad portfolio license agreements and customers ramp-up schedules and the impact on royalty revenues.

CEVA assumes no obligation to update any forward-looking statements or information, which speak as of the respective dates. In addition to the financial results prepared in accordance with the generally accepted accounting principles or GAAP, we will also present certain non-GAAP financial measures today. CEVA's management believes that In addition to using GAAP results in evaluating our business, it can also be useful to review results using certain non-GAAP financial measures. Investors and potential investors are encouraged to review the reconciliation of non-GAAP financial measures with their most direct comparable GAAP financial results, which can be found in the earnings press release issued today. A copy of today's press release for the quarter ending June 30, 2018, and the related financial tables and management commentary, which were included in our current report on Form 8-K filed today can also be found on the Investor Relations portion of our website.

Before handing the call over to Gideon, I would like to remind you that CEVA adopted the new revenue accounting standard known as ASC 606 as of January 1, 2018. Under the new standards, our royalty revenue represents what our customers shipped during the second quarter of 2018 or our best estimates for such shipments. The numbers stated on this call for the second quarter are based on ASC 606, unless otherwise stated. However, as our Q2 2018 financial results are not directly comparable to our Q2 2017 financial results, which are reported under the old revenue accounting standards known as ASC 605, we will also provide you on today's call, our Q2 2018 financial results as reported under ASC 605 to allow for an apples-to-apples comparison on a year-over-year basis. We will have this dual reporting approach throughout 2018 as required by the Financial Accounting Standards Board.

With all of that said, I will now hand the call over to Gideon.

Thank you, Richard, and welcome, everyone. Our second quarter revenue came in at $17.5 million. Licensing and related revenue was $10 million, similar to last year and a quarter ago, reflecting a healthy environment for our AI computer vision and Bluetooth technologies.

Royalty revenue under ASC 606 came in at $7.5 million, similar to last quarter, but below our expectation, attributable to continued and expected weakness of one of our large Chinese handset baseband customers, which I will elaborate on a -- on later in the call. We did, however, experience seasonal unit growth in both handset and consumer product, with non-handset baseband shipment reaching new record level of 88 million units in the quarter. During the second quarter, we concluded 9 deals of which 3 will follow DSP cores and platforms, 1 for our NeuPro AI platform and 5 for our connectivity IPs. 3 of the agreement were with first-time customers and the rest were with existing customer that are expanding their existing businesses or upgrading to a newer product. Customers target application include advanced camera, drones, smart speaker, smartphone, wireless earphones and IoT verticals. We continue to experience good interest in our AI and Bluetooth product, with 5 of our deals in the second quarter being for those 2 product categories.

Let me take the next few minutes to highlight the success factor that distinguish CEVA in the crowded landscape of AI processor and in the cost-driven Bluetooth market. On AI, while most of the AI application are currently cloud-based, there is widespread acknowledgment that AI workload are distinct to move to the devices themselves, or what is referred to as the edge. AI at the edge ensures faster response time and greater privacy by keeping the intelligent processing on the device rather than in the public cloud. Market research firm Tractica forecast that the market for deep learning chipsets will increase from $1.6 billion in 2017 to $66 billion by 2025. Edge devices are expected to represents more than 3 quarter of the total market opportunity with the remainder being the cloud. Smartphones, automotive, surveillance camera and robots are prominent edge categories.

Apparently, the growth opportunity and the market size, the AI space presents attracts companies both from the semiconductor and the semiconductor IP industries to offer variety of AI solution to support the burgeoning need for performance. Against these crowded backdrop, CEVA proposition distinguishes itself in 3 main areas. First is performance. CEVA NeuPro 4000 processor fulfilled the highest AI operations capacity per core among all IP and semiconductor vendors in our targeted market, according to a recent study by The Linley Group microprocessor report.

Second, NeuPro is a unified platform that can combine processing of AI, computer vision and general DSP algorithms. It allows our customer to reuse the same platform for variety of use cases and to complement AI with front-end and back-end computer vision and image enhancements applications. Sales and a key success factor is our deep neural network compiler technology, known as CDNN. CDNN is an incredibly complex software technology that automatically optimizes many different types of neural network to process efficiently, empower and size-constrained edge devices that are enabled by our NeuPro hardware. An equivalent technology called TensorRT offered by NVIDIA, yet its primary use is in the cloud. CEVA is the only company that offers similar features and robustness, targeted for edge devices. We have already signed up to -- up 3 leading customers for NeuPro, who provide us with valuable feedback on this unique offering. We are experiencing interest for pipeline -- from a pipeline of reputable companies, which we will further engage in, as the technology become available for general licensing by the end of this month.

On Bluetooth, according to the recent Bluetooth SIG report in 2018, nearly 4 billion devices will ship with Bluetooth technology. It is expected to grow to other 5 billion units in 2022. The growth will be driven by the latest standard Bluetooth 5 and a recent major addition to the standard, the [mesh] topology. CEVA is the incumbent Bluetooth IP supplier for dozen of customers and clearly there in the Bluetooth dual mode that combine the benefit of Bluetooth low energy with audio capability.

According to SIG forecast, 65% of all Bluetooth shipments by 2022 will be dual-mode, due to the proliferation of smart speaker, Bluetooth audio speaker, earphones, hearing aids and voice-enabled TV remote control. Our newest voice software technology, ClearVox, offers value add in higher royalty level opportunity in conjunction of the Bluetooth and the speakers. We continue to grow our shipments in the Bluetooth space with an all-time high 71 million units shipped in the second quarter, up 57% versus Q2 actual shipments last year.

Let me now provide you with update on our base station RAN customers. Nokia continues to grow the footprint with -- of its AirScale baseband in radio technology enabled by CEVA IPs. Last week, Nokia announced that it has signed a $3.5 billion deal with T-mobile to supply equipment for its 5G network, the world's largest 5G deals announced to date. Nokia has also once designed for 5G at Verizon, AT&T and NTT DoCoMo of Japan and recently signed an important agreement with China Mobile. In its recent earning call, the CEO of Nokia reiterated his confidence for upcoming 5G ramp-up in the third quarter with acceleration in the fourth quarter, but was also prudent about large-scale deployment that is difficult to predict due to the timing of deployment completions.

On ZTE, we are encouraged by the recent lifting of the ban on the sales of U.S. manufactured components to ZTE and the resumptions of their base station operation, which contribute royalties to us. ZTE is a strong contender in the Pre5G and 5G network infrastructure and is determined to resume this position. With that said, we assume it will take a few months for ZTE to rebuild the supply chain and to get back to full productions and shipments. Also, licensing discussion for our more advanced platform, which were put on hold during the ban, will resume shortly, but the timing of licensing closure has been affected by the ban.

As for the handset baseband, in the second quarter, we started to see preparation for sizable ramp-up by one of our customers, with supply baseband processor to our premium smartphone OEM for a flagship launch. In this upcoming product cycle, we are set to benefit from higher ASP due to an increased content. On the other hand, the weakness we experienced in the first quarter in the low tier of the smartphone space with large Chinese based customer of our, continued unexpectedly into the second quarter. We believe, this can be attributed to a share shift at one of large OEMs in which the volume this year will now be split between another customer of ours and a supplier that does not use our technology. We now expect this weakness to continue the rest of the year. So net-net on an annual basis, the expected growth in volume and content associated with the premier smartphone will partially offset the weakness of our Chinese based customers. Yaniv will later discuss the financials implication on our 2018 guidance, which reflect both ZTE and the handset development.

In summary, we continue to progress with our licensing business, driven by healthy demand for our AI computer vision and connectivity product. We continue to strengthen and broaden our technology base and are confident about our strategy and our ability to add more content and value to our customer. On royalties, we are capitalizing on you -- new royalty growth engines notably in premium smartphone and the continued expansion in non-handset segment that will drive substantial sequential royalty revenue growth in the second half of the year and beyond.

With that said, let me turn the call over to Yaniv, to discuss our financials and guidance.

Thank you, Gideon. I'll review the operations for the second quarter of 2018. Revenue for the second quarter based on ASC 606 was $17.5 million. The revenue breakdown is as follows.

Licensing and related revenue was $10 million, reflecting 57% of our total revenue, only 3% lower as compared to the second quarter of 2017. Royalty revenue was $7.5 million, reflecting 43% of our total revenue, a decrease of 26% year-over-year basis compared to $10 million for the second quarter actual shipments that were reported in the third quarter of 2017, following the new rules under ASC 606. Quarterly gross margin was 89% on U.S. GAAP basis and 91% on non-GAAP basis. Non-GAAP quarterly gross margin excluded approximately $167,000 for equity-based compensation expenses and $183,000 for the impact of amortization of the acquired intangibles of our investment in the narrowband IoT technologies last quarter.

Total operating expenses for the quarter were just below the mid-range of our guidance at $18.2 million. OpEx included an aggregated equity-based compensation expense of approximately $2.7 million and $0.1 million for the amortization of acquired intangibles of RivieraWaves. Our total operating expenses for the second quarter, excluding these 2 items were $15.4 million, slightly above the midrange of our guidance. U.S. GAAP loss and diluted loss per share were $2.1 million and $0.09, respectively. Our non-GAAP net income and diluted EPS for the second quarter of 2018 were $0.9 million and $0.04, respectively. These figures exclude equity-based compensation expenses in the amount of $2.8 million and the impact of amortization in the amount of $2.3 million.

Our second quarter 18' financial results under the old 605 rules compared to the second quarter of 2017 were as follows: Total revenue was $16.6 million, U.S. GAAP loss $3 million and $0.13 loss per share and the non-GAAP net income and EPS for the second quarter were 0 -- cents and dollars, respectively.

Other related data. Shipped units by CEVA's licensees during the second quarter of 2018 were approximately $222 million, up 13% sequentially, and down 11% for Q2 actual shipments reported in the third quarter of 2017. Of the approximately 222 million units shipped, 134 million or 60% were for handset baseband ships, reflecting a 10% sequential increase and a 29% decline on a year-over-year basis. A non-baseband -- handset baseband volume shipments reached a record high of 88 million units, up 19% sequential and 44% on a year-over-year basis. Bluetooth shipments continue to be strong.

As for our balance sheet. As of June 30, 2018, CEVA's cash, cash equivalent balances, marketable securities and bank deposits were approximately $173 million. During the second quarter, we paid ASTRI 2 more payment milestones of just less than $1 million for the new narrowband IoT technologies, as we discussed on our prior earnings call. We continue to be active in our buyback program, repurchasing approximately 270,000 shares during the second quarter and an average price of $33 per share for approximately $9 million. During the second quarter our Board of Directors approved the expansion of the existing buyback plan and at the end of last quarter, we had the total of 700,000 shares available for repurchase. Looking back, in the last 10 years of our buyback activities, we repurchased 6 million shares for almost $100 million.

Last, our adjusted ASC 606 DSOs for the second quarter of this year were 48 days, down from the prior quarter of 62 days. During the second quarter, we generated $2 million from net cash from operations. Our depreciation was $0.6 million and repurchase of fixed assets was $1.8 million, higher than the norm, due to additional new EDA tools for R&D design teams. At the end of June, our headcount was 325 people, of which 261 are engineers.

Now for the guidance for the rest of the year. On royalties, we are lowering our 2018 annual royalty guidance to a 10% decrease from the 2017 level. With that said, we expect a substantial increase in royalties in the second half of the year, with more than 50% sequential increase for the third quarter. Moreover, royalties are expected to turn to year-over-year growth in the second half of 2018. On licensing and related revenue, we continue to experience healthy demand for our products but it slightly reduced our expectations for the year due to the ZTE deal push-out, which Gideon elaborated on.

Our revised revenue for the year is expected to be in the region of $80 million, split fairly evenly between licensing and royalties, plus or minus few percentage points either way. Specifically for the third quarter of 2018, gross margin is expected to be approximately 91% on GAAP and 92% on non-GAAP basis. Our overall OpEx is expected to be slightly lower than the second half -- at the second quarter OpEx in the range of $17.5 million to $18.5 million. Of the anticipated overall, OpEx for the third quarter, $2.4 million is expected to be attributed to equity-based compensation expenses and $0.2 million for the amortization of acquired intangibles. So our non-GAAP OpEx is also expected to be similar to the second quarter level in the range of $14.8 million to $15.8 million. Net interest income is expected to be approximately $800,000. Tax rate for the third quarter 17% on GAAP, and 11% on non-GAAP basis. And last, our share count for the third quarter is expected to be approximately 23 million shares.

Brandon, you should now open the floor to the Q&A session, please.

(Operator Instructions) Our first question comes from Gary Mobley with Benchmark.

I'm not sure what it is for you guys, but with respect to the 5G infrastructure royalty opportunity based on your existing licensees, Nokia and ZTE, can you give us an update as to your expectations once those guys are up at a full run rate? And as well, what your outlook for 2019 may look like, whether it be on an absolute basis or on a delta compared to 2018?

Gary, it's Gideon. Let me speak in general about it. When it comes to Nokia, we can only echo what they're saying and they're speaking about Q3 and Q4 deployment, starting Q3, with larger extent in Q4. I believe it's going to be on the T-Mobile network. We'll have to see how this looks like, but from our standpoint, just wait and see. When it comes to ZTE, I'll elaborate it. They are coming out of the ban, the ban was not just -- not cause just shutting down their operation, basically their supplier stopped manufacturing chips and in some cases even the people were moved to other project. So this takes time to rebuild. So we are thinking about Q4 to start seeing a nice recovery and we bake some amount into the Q4 royalty. So without speaking of waiting, I think, when it comes to see a substantial, we need to speak about 2019, when ZTE will be behind that and they will be behind that and Nokia are starting leveraging on the wins that they had.

Okay. And based on the size of the market and your share, which I believe is 50%, what do you think the full run rate royalty revenue could be?

I don't think we changed the -- our longer-term forecast. I think we still believe that this is a significant royalty opportunity and based on all the data that you just said, the market size and the number of chips in its base station, as it could be north of $20 million. We just want to see those first royalty reports. We -- as you know, we did get the 1 or 2 from ZTE and then the ban kicked in. So we want to see that get back. And that the opportunity could be way over $20 million, if they do it right and both of these customers are in full production in the first quarter of 2019.

Okay. As a follow question, focusing specifically on the mobile handset baseband process insight. It sounds like your royalty rate per unit may increase at the Tier 1 smartphone OEM that's based in the U.S. as your shares -- as you gain share there in the next product cycle. But on -- first of all, can you confirm that your 4G royalty rate, as a result of that, might increase in the second half of the year? And then with respect to overall handset market share, just given the moving pieces -- do you think your market share will shrink in 2018? And based on your share gains at the Tier 1, do you think your share can increase in 2019?

Yes. So let's start with the share. The first 2 quarters of the year, we were 23% and that reflected first, very soft Chinese market overall in the first quarter, if you remember. That caught not just us, but the whole semiconductor market by surprise. What was unexpected for us, was that one of our key customers in that space continued that weakness into the second quarter and didn't show any seasonal increase like we have seen with other customers in the handset space. So we are -- we were stuck in the second quarter with the same 23% market share. The second half of the year is -- should look different. And right now, we have not baked in growth for that Chinese player for the second half or at least not a significant one, because we want to see how they deal with the problems and hopefully, they could get over them and win some new sockets. With that said, and when we take the premium handset that should be launched in the third quarter, we will see those 2 elements that you mentioned, 1 is the higher raised fee for overall the smartphone space for us because of that ramp, and second, an improvement in the overall market share that we'll have -- it won't go back to 40%, but it should be better than our 23%, with higher dollar content. So the way we are seeing the model for the second half of the year, that with the softness with the Chinese and the strong volume and ASP increase from the other handset baseband player that is ramping up, we will -- we hit our numbers with only maybe $2 million net. So that's part of the reason for reducing our annual royalties, but the manifest is not as bad and hopefully, we could see some better moving pieces for the -- from our Chinese large customer.

Our next question comes from Mike Walkley with Canaccord Genuity.

Just building on that question, can you maybe just walk us through on what's your lower royalty guidance through the year. Just how much was that related to maybe ZTE? How much to the weaker smartphone market? And then how much to the Chinese OEM weaker than you expected?

Sure. So let's talk and look how we started the year and what we wanted to reach. We were talking about $92 million with licensing revenue flattish at the $43 million and about 10% growth in royalties. The first half, we were down from those expectations just in royalties by about $5.5 million. $1 million was the ZTE ban in Q2, which we shared with you and the rest of the first half $4.5 million were equally distributed between a very soft Chinese market overall, not just for our customers, but for everybody, including Qualcomm in China in Q1. And the -- another half of that $4.5 million of about $2 million, $2.5 million was associated with the softness with our specific Chinese vendor in the second quarter. So overall, that's $5.5 million, that's behind us, this is the first half of the year and there is not much we could do about it. ZTE, we're happy that the ban has been lifted, it took a while, it took close to 4 months. If it would be quicker few weeks, we would be -- they would be back in production much earlier. So when we start Q3, we are taking into account maybe about $1.5 million associated with that and partially into Q4 as well. And we think that by the end of Q4, maybe the last month of Q4, they'll be back in production and we'll see some royalties coming from Q4. So that by itself, just from the ZTE ban in the second half, you're probably looking at $2 million. If we are talking about ZTE, Gideon mentioned that they were also evaluating some of our newer technologies for next generation and should design with us. And so we're taking offers about $2.5 million in the licensing side and trying to build a $40 million, $40 million to $41 million type of base. So ZTE overall, in the second half of the year, I would say almost half, half between royalties and licensing, taking overall about $4.5 million. So $5.5 million from the first half of the year and $4.5 million brings us to $10 million, the last $2 million is what I just answered earlier that, that's the net effect of the weaker Chinese customer versus a very strong offset, but not complete offset of the new smartphone vendor, the premier smartphone vendor that is going to use us with more content and higher volume this year versus last year. So we've almost been able to offset that handset space. But we're off about $2 million, as we see it right now for the second half. There are moving pieces like other non-handset baseband growth that we've seen in the first half of the year and that should continue in the second half. And the Nokia of course, as Gideon said, Nokia is still and has been throughout the year, sort of, a moving target for us. We don't know specifically how big that could be in the near term, which is Q3 and Q4. We have baked something in, but far less than what we believe could be the full extent of Nokia ramp-ups with all these design wins that they are talking about, NTT DoCoMo, China Mobile, T-Mobile, which is just around the corner, and we just need to follow these guys and see how fast they get -- can get deployed and in what city. So if you see any deployments around your city, give us a call and we will monitor the royalty reports for the upcoming quarter.

And my follow-up question, just on licensing deals, absent in ZTE is everything, kind of tracking has progressed and maybe, could you highlight some of the growth opportunities you're seeing in licensing, both the rest of this year and into next year, including any update on automotive opportunities?

Yes, it's Gideon. So the 2 high running product in terms of licensing today in ZTE is AI. Keep in mind that when it comes to AI, we are not in full scale in a -- in going out to customers until the end of this month. We set a target for the AI to sign up one or two lead customer that help us to polish the product before going out to the market. We were lucky to get 3, this is very helpful for us but by the end of the month, we going to be there with production, where the technology. And I believe we'll see more customer that are more risk averse. So that's a -- I rather think is extremely successful, is the connectivity in general. Bluetooth, right now and we see Wi-Fi coming in the same context, every connected device, need the Bluetooth and Wi-Fi. Bluetooth, huge volume, I -- it's a 4 billion unit this year, lower ASP -- royalty ASP and licensing Wi-Fi, it's also a very big market and here we get higher license fee, it could be 2x, 2.5x of Bluetooth and of course, higher royalty. So these are the 2 high runner products for this year. For next year, I -- the way I see it, is that, we're going to see more 5G coming, both in base station run, there are few customers that we would like them to join us. We want to be a monopoly there, we want to dominate the space and we are not far apart for that. And there will be 5-- we see 5G handset maker, right now, well the 5G designs are more of a show kind of a concept level design, it's not mass production. This is not -- in such case, we don't usually participate, but when people go to mass production and wants to get the power metrics and the capability that we are offering, in terms of software programmability, then they call us. So 5G and -- both in base station, I think we're going to see more and more in 2019.

Thanks, Mike. One more thing I forgot to mention, that when we jumped from this very low -- relatively low first half of the year to Q3, I mentioned that we're going to see like 50% growth -- sequential growth in royalties. So we're picking up, of course, from a low-level, that we didn't want to be there, but it seems that we're back on track to almost the same expectations that we had for the second half of the year. And hopefully, some of these things that talked -- we talked about with base stations and even hopefully, the Chinese baseband player could do a bit better than we have -- our planning for the second half.

And our next question comes from Joseph Wolf with Barclays.

I was hoping we could -- you could give a little bit more detail on the units of the non-baseband, $88 million in the quarter, $160 million or so for the first half. Where does that go in the second half of the year, just in terms of number of units that you guys are in? Is there another big pick up, 25% in Bluetooth and what's in the other right now? In terms of what is exactly -- what is shipping that's not Bluetooth, that's giving that 17 million unit number?

So it's a collection of product. We will not see or will seem minor contribution for the base station, that in unit terms the contribution is minor, but the dollar terms are high. But when it comes to other, that are non-Bluetooth, it's a collection of different chips, different markets, could be computer vision, could be audio, could be narrowband IoT, could be industrial application. We just -- it's just a portfolio of things. And many customers -- some of them are new and once we start seeing them coming out, we'll analyze more and project the next level. But for now, we want them to go out of the door, go to production chip units and then help them to get to masters as well.

The first success, Joseph, is really the Bluetooth that started off maybe 4 years ago with us. And if you remember from 0 royalties, when we got into this space, we are in the run rate of 71 million a quarter, just Bluetooth devices, all over the place from hearing aids to speakers, to wire, to sport wristbands, to camera and functionalities and lots of other things, I hope it will have. This is the plan, as Gideon said, to have others like in the vision, like in the audio follow up with much higher volumes, the opportunity is there and the design wins are there. For now, the quickest ramp-ups after so many designs that we have in the connectivity side, we also see -- already see 2 new customers in Bluetooth and Wi-Fi for the second half of this year. So the volumes, absolutely, should continue to go up, first from seasonality pre-Christmas point of view and second, from the number of customers that are finally getting into production. And we've been waiting for a while, and the Bluetooth and Wi-Fi are working very nice within 3 years, maybe after the design starts, sometime a bit earlier, we are seeing them deployed into different end markets.

Joseph, just one more thing that I want to shed light on Bluetooth. Bluetooth is huge market. I mentioned in my prepared remarks, 4 billion units just this year. And I suggest to watch how Nordic and Dialog are doing. They're all in -- make their money out of Bluetooth and they are speaking about in the chip side, high [terms] of percentage of growth in year-over-year in unit terms and revenue. So it's a very big space and if you do it right, you can flourish them.

Okay, that's helpful. And then, if I look -- there was 1 handset related license of work that you guys did in the quarter. Is that a 5G related license? And is there any reason to believe that the 2019 licensing revenue will see a lot of -- or new kinds of designs for handset coming back?

Designs for handset could come in 2 -- from 2 angles; one is, incumbents and they are people that use us and will continue to use us, and there are people that don't use us and we believe, we have a proposition for 5G an entry point for them. And I wouldn't exclude any incumbent in the cellular market today, in the handset market, saying, no, this guy is [very much] and he has everything. The other part and there are newcomers that are into the space. There is 1 in China, I don't want to mention the name because I'm not sure that we already announce it, but is a strong -- could be a strong contender in this market. And we see few others that are looking into the space, not necessarily because of handset, because 5G spend is beyond this automotive and IoT in general. And that's also a big market. So in terms of license in 5G, we are speaking now in a bigger term, in terms of available licensees.

Okay. And then just finally, with all the buyback, can you just give an update on whether that means that you're seeing fewer M&A opportunities? Or you just -- there is no appetite? Or it's -- you've got the right strategy and aren't looking around that much right now?

So let me respond on the M&A strategy and then Yaniv can refer to the buyback. When it comes to M&A, we are prudent. It's not that there are a hundreds of companies that we can buy. We did one very successful acquisition and we want to continue using it. So we don't -- right now, we don't have something specific that we have in our radar. And we are not rushing to do it. We have an organic strategy that has enough engines to flourish.

Yes, nothing really to add. The buyback is here to support the stock and our belief in the longer term, which is not too long, it's now the second half of the year, which is much better than the first one and much better than any second half we had so far. And it's just for the time being, it's a good use of the cash and IP mall should generate cash, overall. So this is just something that we're doing in the meanwhile, until we find the right M&A target, the right technology that we want to have. So I don't think those 2 contradict, we are not talking about $100 million buyback program but a smaller 1, as we go along.

Our next question comes from Matt Ramsay with Cowen.

I guess, my first question is on the base station market. Gideon you made -- mentioned in your comments just now that you guys are looking for eventual full coverage of the base station space with your technology. obviously, you've announced a couple of big OEMs. Maybe you could just give us an update on where you stand with, sort of, the other big 3? Obviously, Ericsson, Samsung are things that maybe you've hinted at before, that there is some dialogue, but the folks at Huawei do their own DSPs, typically. So maybe you could give us a little of update on just generally where discussions are with the rest of the market? And if there were some royalty deals to get signed or some licenses to get signed rather, like how long should we think about the time line for royalties coming out of the rest of the base station players?

So out of the 3 names that you mentioned, one is already working with us, one is in discussions and one, there is no discussion. So I'm sorry, I cannot elaborate more, but we're the big shot there, because we are the only guy in town that can provide such complicated technology for 5G. Now beyond that, people are speaking about base station and think about the incumbent there, but there are all those small sales in fixed wireless and we add into this one all the enterprise, so all the Wi-Fi access point. For us, it's the same technology base. And then you start speaking about different name, you speak about networking guys in here, the opportunity is huge. And we are here doing real steps and progress with few of the large guys there.

No, I appreciate the sensitivity around some of the ongoing discussions. I guess as my follow-up questions, I had 2 questions or topics I wanted to sort of ask about in the smartphone space. I guess, on the more challenging side, obviously, there's some things going on with Spreadtrum here. I guess my question around that is, what was new and different for this quarter's update? And maybe you had updated us on 3 months ago, I think it was fairly well known in the industry that MediaTek was being pulled into Samsung at the low end and that was going to be a challenge. So I guess maybe, Gideon, first of all, what's new and different that's happened with that particular customer? And second, on the positive side, Qualcomm's all but admitted that Intel is going to be 100% share of the flagship iPhone going forward. Maybe you guys could talk about what your share has been or what you estimated to have been within the iPhone on the modem side in this last cycle? And what kind of an uptick we could look at and particularly, sounds like you mentioned that ASPs or royalties per unit could be a bit higher, with that customer going forward. Any commentary on just magnitude of what we should think about from a royalty shift would be really helpful.

So without going to the specific details that you would like to see, I'll give you just the envelope of that. So when it comes to the Chinese customers, the first quarter was all around weakness in the market, there was not any -- there are different segments, feature phone and 3G and LTE and all of them were ugly. And it was correlated, what's going on, on the market overall. What I already said in my prepared remark, is a specific OEM, large OEMs that Chinese customer had a certain share and a very large share, and now it's being splitted between another customer of ours, so we are not -- with this specific OEM, we're not completely out. We are left with substantial share, but not the same share that we do. And that's the sort of the revise -- the update in the guidance that we will do it. With that said, cellular market is a very dynamic market and this may answer your second question. It work on a 1-year cadence, and if you lose a socket, in a big socket or small socket, usually its take you -- you are out of it for the whole year, but then you have another chance to get the next deal. So specific to these Chinese customers, where we see some signs of not regaining share, by gaining share with other OEM on the account of the incumbents there. So it's a big fight. And we need to take it over there. It is what it is. But this company -- this Chinese company was a competent company and will be a competent company, and there is lot of support there in China to -- not to drag into the process -- to fix the problem. It's not -- I wouldn't -- you lost a design and then you give another shot and you do it, but that it is what it is, that's how this market works.

Got it. And any update on sort of the big baseband customer in the U.S. would be really helpful going into 2019.

We're waiting for the same data. I think the public information about the ramp-up with T-Mobile is there. We need to either see the sign that it's happening on the streets or get the royalty report and then figure out the volume and magnitude. So this is something that we have been following and updated -- outstating that we don't have any special insights, other than the public announcements that these guys are talking. And they have been talking a lot on the base station side in the U.S. The handset side in the U.S. I think we mentioned, higher speed, higher volume, higher share in that socket from the third quarter, which all are benefiting factors to offsetting almost completely, but other than $2 million for the second half for this year.

Our next question comes from Suji Desilva with Roth Capital.

For the smartphone baseband, the folks that typically haven't used CEVA, looking ahead to 5G, what's the likelihood of some of those baseband vendors would consider using CEVA? And if they were to, would it be a mixed, some platforms being CEVA, some non-CEVA? I want to understand the handicap there.

So if you ask us about OEMs because when it comes to our relationship with the baseband semis, this is our ongoing relationship and our strategy there is to convince them to upgrade to a newer technology and we do have a lot to offer them. When it comes to our customer relationship with OEM and that's what I tell Matt in the other question, it's really very dynamic. And we have to take it on a yearly basis. So in any -- one of your large customer can win a socket, like it's happened today -- this year's premier cycle and you can lose in another large OEMs to do -- it's a big market. Today, it's more fragmented, it's not just Apple and Samsung and then you have Oppo and Vivo and Xiaomi and ZTE. And all of them can create tens of millions or get even to hundreds of million units a year. So if you lose 1 customer, you can go to -- you can win another customer. And by the way, that's what happens to the Chinese customer. We are seeing in them, winning sockets in other customers. It will come to -- into play in 2019, when these customers will refresh their portal. So we -- other than this, we don't have what to say, other than this specific Chinese customer is a viable player and continue to be a viable player, and what we lost today, we'll win somewhere else.

Okay. And then a question on the wireless infrastructure side, a lot has been asked. But in terms of Nokia ramping up here, as you come closer, it's the second half and you're seeing some visibility. Is the linearity playing out as you expected? Is there any difference from what you expected? And how it's playing out -- are the units for the second half '18 coming in as expected above or below what you had forecast?

No below, for sure below because, as I -- as we mentioned, we still don't have the visibility for those ramp-up's that Nokia just a week ago, talked about in Q3 and Q4 deployment in T-Mobile. So as soon as we have that data, we'll update. We thought it could be a bit earlier, in the beginning of the year, we didn't know anything about ZTE band. So from both base station customers, it's been taking a bit longer. But I think all the obstacles are off and out of the way and ZTE can now go and continue their manufacture, and they already was a royalty contributor. And Nokia is just about to start, so we hope it's a good start for them and we wish them enough luck, because that's what they are saying, that from Q3, they're starting to deploy it in different networks.

Okay, great. And then my last question, it sounds like the AI opportunity is starting to come in from the license side at least. Can you remind us what the royalty opportunity in AI is, when it kicks in? What the ranges of the potential royalty there is?

Yes, it could be quite significant, it depends really, that what type of AI, it's a long discussion. I think our time is out, but AI could go into so many different elements. We talked in the past about seeing them with vision, next to audio products, in the automotive, we are talking with few vendors. So it could be anywhere from a vision type of process to anywhere to up to a $1 type of ASP, it depends really on the end market. It's a very lucrative product line for us, both licensing and the royalty potential from it.

This concludes our question-and-answer session. I would like to turn the conference back over to Richard Kingston for any closing remarks.

Thanks, Brandon. And thank you, everyone, for joining us today and your continued interest and support of CEVA. We will be attending the following upcoming events and invite you to meet us there. Tomorrow, we'll be at the Oppenheimer 21st annual technology, Internet & Communications Conference in Boston. On Thursday, we'll be at the Canaccord Genuity, 38th Annual Growth Conference in Boston. We'll be on September 5 at the Roth Internet of Things Corporate Access Day in San Francisco. Also on September 5, we'll be at the Citi 2018 Global Technology Conference in New York. And on September 12, we'll be at the Deutsche Bank Technology Conference in Las Vegas. Please visit the investor section of website for further information on these events and other events we will be attending. Thank you and goodbye.

The conference is now concluded. Thank you for attending today's presentation. You may now disconnect.