Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

stds-802-16-tg4: TG4 PHY STRAWMAN COMMENT:



Title: TG4 PHY STRAWMAN COMMENT:

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
4

[Section #]
3.

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
E

[Detailed Description of Proposed Insertion, Deletion, Change]
Replace DFT with FFT.

DFT is more general but FFT is typically used in the OFDM, 802.11 and 802.16 slangs. Also, only FFT-size is limited to powers of 2.

[Reason for Edit]
Consistency

--------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
4

[Section #]
3, 4, 5.

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
E

[Detailed Description of Proposed Insertion, Deletion, Change]
Replace: upstream, US, downstream and DS with uplink, UL downlink and DL.

[Reason for Edit]
Consistency with 802.16 terminology.

---------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
36, 37, 47

[Section #]
5.2, 5.3, 9

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
E

[Detailed Description of Proposed Insertion, Deletion, Change]
Integrate (optional) features from these sections in section 3.
Get Yossi and Tal to help Nico with his huge section.

[Reason for Edit]
Clarity.

--------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
4

[Section #]
3

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
E

[Detailed Description of Proposed Insertion, Deletion, Change]
Reorganize the section as follows:

3.1 Frame structure (both OFDM and OFDMA modes) as in 802.16 and the mesh mode

3.2 OFDM modes
3.2.1 Burst(s) structure
3.2.2 Carrier allocation and modulation for short preambles
3.2.3 Carrier allocation and modulation for long preambles
3.2.3.1 64-FFT
3.2.3.2 256-FFT
3.2.4 Carrier allocation for data symbols
3.2.4.1 64-FFT
3.2.4.2 256-FFT
3.2.5 Pilot modulation (data symbols)
3.2.5.1 64-FFT
3.2.5.2 256-FFT
3.2.6 Optional subcarrier based polling
3.2.5.1 64-FFT
3.2.5.2 256-FFT

3.3 OFDMA mode
3.3.1 DL
3.3.2 UL
3.3.3 Ranging
3.3.4 BW request
   
[Reason for Edit]
Clarity.

---------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
16

[Section #]
4.12

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
T

[Detailed Description of Proposed Insertion, Deletion, Change]
Fixed initialization of the scrambler is not compensated by the length of the scrambling code. Consider using a simpler, shorter scrambler that is initialized either as in 802.11a or with some variable but known information as frame number.

   
[Reason for Edit]
Improving performance, reducing cost.

---------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
21, 17

[Section #]
4.2.2.2.1, 4.2.1.1

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
T

[Detailed Description of Proposed Insertion, Deletion, Change]
At Tx, the proposed tail biting techniques requires that the tail of the block is known before starting to process the block. This requires additional storage space and may introduce an additional delay in Rx/Tx switching.

We can use an alternate tail biting technique: the convolutional coder is initialized with the first 6 bits of the block, it processes the rest of the   block and then the 6 bits used for initialization. This way, no additional delay or storage space is required either at Tx or Rx.

Such scheme can be extended to large blocks without any loss and therefore can be used also with the pure convolutional encoder in section 4.2.1.1.

[Reason for Edit]
Improving performance, reducing cost.

---------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
22

[Section #]
4.2.2.4

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
T

[Detailed Description of Proposed Insertion, Deletion, Change]
I see no reason to use an interleaving algorithm with variable computational effort per step. I suggest using a simple, deterministic method as in 4.2.1.2 that can be applied to any multiple of 16 carriers.

[Reason for Edit]
Improving performance, reducing cost.

---------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
34

[Section #]
5

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
T

[Detailed Description of Proposed Insertion, Deletion, Change]
Replace DVB-T constellations with those used in 802.11a or 802.16.1 PHYs.

[Reason for Edit]
Consistency with our 802.16b PAR.

---------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
37

[Section #]
6.1.2

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
E

[Detailed Description of Proposed Insertion, Deletion, Change]
Describe FCC limits in amplifier and radiated power (EIRP) to clarify the 6dB limit on antenna. The out of band emission must be also in EIRP.

[Reason for Edit]
Clarity.

---------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
41

[Section #]
6.4.3

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
T

[Detailed Description of Proposed Insertion, Deletion, Change]
Remove section 6.4.3. Since the standard defines the modulation, there is little somebody can do to improve interference in demodulation. Thus, there is no way somebody can improve co-channel interference properties.


[Reason for Edit]
Clarity, simplifying test procedures.

---------------------------------------------------------------------------------

[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
41

[Section #]
6.4.5, 6.4.7

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
T

[Detailed Description of Proposed Insertion, Deletion, Change]
Remove these two sections. Minimum and maximum Rx signal define the quality of the receiver.

[Reason for Edit]
Clarity, simplifying test procedures.

---------------------------------------------------------------------------------
[Submitter's Last Name]
Sarca

[Submitter's First Name]
Octavian

[Starting Page #]
42

[Section #]
6.6.3

[Starting line in Section #]
1

[(T)echnical for Content-Related Material; (E)ditorial for typos, grammar, etc.]
T

[Detailed Description of Proposed Insertion, Deletion, Change]
Replace 24Vdc by 48Vdc as it the standard power supply for most industrial equipment and carrier class telecom products.

Add 230V for European equipment.

[Reason for Edit]
Generality.

---------------------------------------------------------------------------------


Octavian Sarca, PhD
Director, Hardware Development
Redline Communications Inc.
90 Tiverton Crt. #102
Markham, ON, L3R 9V2
E-mail: osarca@redlinecommunications.com
Tel: (905) 479 8344
Fax: (905) 479 7432