Synchronization signals for narrowband operation

1. A method for wireless communications by a first type of user equipment (UE), comprising:

detecting, within one or more narrowband regions, a primary synchronization signal (PSS) generated utilizing a first code sequence and a cover code applied to the first code sequence over a first number of symbols within one or more subframes, wherein the cover code comprises a binary sequence;

performing a time offset estimation and a frequency offset estimation, based on the PSS, wherein the time offset is uncorrelated with the frequency offset; and

performing an initial time synchronization and a frequency synchronization based on the PSS.

2. The method of claim 1 , wherein performing the time and frequency offset estimations comprises performing maximum-likelihood time and frequency offset estimations based on the PSS.

3. The method of claim 1 , wherein performing the time and frequency offset estimations comprises performing a maximum-likelihood frequency offset estimation of a normalized frequency offset having a fractional part and an integer part.

4. The method of claim 3 , wherein performing the frequency offset estimation comprises estimating the integer part by a cross correlation of received and transmitted PSS pilots in frequency domain.

5. The method of claim 3 , wherein performing the frequency offset estimation comprises estimating the fractional part based on the time offset estimation.

6. The method of claim 3 , wherein:

performing the time offset estimation comprises normalizing and performing a peak search of the time offset; and

performing the frequency offset estimation comprises:

aligning symbol boundaries of the PSS;

correcting the fractional part by phase differentiation; and

correcting the integer part by frequency domain correlation based on the corrected fractional part and symbol boundaries.

7. The method of claim 1 , wherein performing the frequency synchronization comprises:

performing a fractional frequency offset correction, from the PSS, based on autocorrelation in time domain of the time synchronized PSS or based on multiple hypothesis testing and frequency binning; and

performing an integer frequency offset correction, from the PSS, based on cross correlation in frequency domain between the PSS and a local replica of the received signal.

8. The method of claim 1 , further comprising:

detecting, within the one or more narrowband regions, a secondary synchronization signal (SSS) generated based on a second code sequence over a second number of symbols within the one or more subframes; and

performing at least one of fine time synchronization or fine frequency synchronization based on the SSS, to refine the initial time and frequency synchronization.

9. The method of claim 8 , wherein performing the fine time synchronization comprises:

performing a timing offset correction based on a maximum likelihood (ML) estimation by coherently combining all co-phased symbol pairs according to a pattern of the cover code, wherein a range of the combining spans across the one or more subframes; and

normalizing a cost function by a power of the PSS sequence to reduce sensitivity to signal to noise ratio (SNR) levels.

10. The method of claim 9 , wherein the combining of the co-phased symbol pairs comprises:

performing autocorrelation between delayed samples of the PSS, wherein a lag size of the delayed samples is a configurable parameter based on PSS time domain resource allocation.

11. The method of claim 9 , wherein a pattern of the cover code is selected to shape the ML estimation of the timing offset into a desired form.

12. The method of claim 1 , wherein the binary sequence is a computer generated binary sequence.

13. The method of claim 1 , wherein PSS and SSS are detected in non-overlapping resources.

14. The method of claim 13 , wherein PSS and SSS are detected in separate subframes.

15. The method of claim 1 , wherein performing the fine timing synchronization comprises performing cross correlation in time domain on the SSS.

16. The method of claim 1 , wherein the first number of symbols are mapped to resource elements that have a same subcarrier spacing as resource elements used to communicate with a second type of UE that communicates on the wider system bandwidth.

17. The method of claim 16 , wherein the PSS and SSS are transmitted using resource elements not used for communicating with the second type of UE.

18. The method of claim 1 , wherein:

the first number of symbols are within a first subframe; and

the second number of symbols are within a second subframe.

19. The method of claim 1 , wherein the first code sequence is generated using at least one of a generated sequence (CGS), a modified Zadoff-Chu sequence with a reduced alphabet, a Walsh code, a Barker code, or a truncated Zadoff-Chu sequence.

20. The method of claim 1 , wherein the second code sequence comprises a Zadoff-Chu sequence, a computer generated polyphase sequence, or a truncated Zadoff-Chu sequence.

21. The method of claim 1 , further comprising determining, from the SSS, a cell identification (cell ID) and additional system information.

22. The method of claim 21 , wherein the additional system information comprises at least a subframe ID.

23. An apparatus for wireless communications, comprising:

at least one processor configured to:

detect, within one or more narrowband regions, a primary synchronization signal (PSS) generated utilizing a first code sequence and a cover code applied to the first code sequence over a first number of symbols within one or more subframes,

perform a time offset estimation and a frequency offset estimation based on the PSS, wherein the time offset is uncorrelated with the frequency offset, and

perform an initial time synchronization and a frequency synchronization based on the PSS.