Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs) Submission Title: [Response to optional CoU and CS pulses related comments] Date Submitted: [March 8, 2006] Source: [Huan-Bang Li, Kenichi Takizawa, Yuko Rikuta, Shinsuke Hara, Tetsushi Ikegami, and Ryuji Kohno] Company [National Institute of Information and Communications Technology (NICT)] Contact: Huan-Bang Li. Voice:[+81 46 847 5104, E-Mail: lee@nict.go.jp] Abstract: [Suggested response and remedy on comments related to optional CoU and CS pulses ] Purpose: [To help the resolve no-vote comments for 15.4a draft] Notice: This document has been prepared to assist the IEEE P802.15. It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein. Release: The contributor acknowledges and accepts that this contribution becomes the property of IEEE and may be made publicly available by P802.15.

Response to Optional CoU and CS Pulses Related Comments Huan-Bang Li, Kenichi Takizawa, Yuko Rikuta, Shinsuke Hara, Tetsushi Ikegami, and Ryuji Kohno National Institute of Information and Communications Technology (NICT)

TR Comment No.441 Comment The y scales of the graph are not defined. What are these? Response Accept. The y scales of the graph is amplitude. Remedy Add ‘amplitude’ to the y axis of the graph. (Fig. 27h) amplitude

TR Comment No.442 Comment Response The Greek letter mu is already in use in this standard for a different quantity. Pick a different letter. Response Accept. Remedy Replace ‘mu’ with ‘beta’ (Page 37, Line 4 - 5 and Table 39I) where r(t) denotes the mandatory pulse shape given in Section 6.8a.3.3.2 and the chirping rate (chirping slope).  (slopes) (slopes)

TR Comment No.443 Comment Response “will remain the same” is not strong enough. Change to “shall be the same”.. Response Accept. Remedy Change as suggested (Page 36, Line 7-8 Page 37, Line 10-9 from bottom) all modulation specifications will remain the same as they are for the mandatory pulse shape all modulation specifications shall be the same as they are for the mandatory pulse shape

TR Comment No.444 Comment Response 'tau(f) is not in units of seconds/Hz, it is units of seconds. If, on the other hand you are describe a constant tau that is multiplied by frequency as opposed to a function tau that has frequency as a parameter, then the constant would indeed have units of second/Hz 'Remove references to tau(f) and replace with tau * f Response Accept. Remedy Change as suggested (Page 37, Equation o-2 and the line under it) See next page

TR Comment No.444 (Continue) where, represents the group delay where, represents the group delay

TR Comments No. 610 and No.611 Comment The standard as produced is really complicated for a sensor application. Furthermore, optional pulse shapes and the like will thwart true standardization, as equipment built with the different pulse shapes will not be compatible. To simplify, I suggest eliminating all optional pulse shapes. This applies both to the chaotic pulses and the continuous time pulses. Response Reject. The reasons are given in following. One can select to install the mandatory pulse only if this meets the requirements. Optional pulse shapes only need to be installed when one want to take the advantages provided by the optional pulses. Inter-operability is guaranteed because FFD keep listening with the mandatory pulse even when operated in optional modes.

TR Comment No.762 Comment Interoperation of chirp-capable nodes with non-chirp capable nodes needs to be clarified. It appears that the two modes are completely incompatible, creating what is essentially yet another alternate PHY. Response Reject. The reasons are given in following. Interoperation is guaranteed because FFD keep listening with the mandatory pulse even when operated in optional modes. When both chirp-capable nodes and non-chirp capable nodes exist in a piconet, the piconet can only operate with the mandatory pulse.

TR Comment No.865 Comment These pulses need to support reception by the non-coherent receiver (need to have same pulse or symbol duration and bandwidth, etc). Response Accept in principle. An option will only be validated in a piconet that all devices are installed with this option and the coordinator approves the use of this option. Default demodulation is coherent. To make the CoU option more feasible, the chirp slops in Table 39I are modified. See next page.

TR Comment No.865 (Continue) Table 39I CoU channel slopes Table 39I CoU channel slopes Channel number (slopes) Ch.1 500MHz/100ns Ch.2 -500MHz/100ns Ch.3 1GHz/200ns Ch.4 -1GHz/200ns Ch.5 1.4GHz/300ns Ch.6 -1.4GHz/300ns Channel number (slopes) CCh.1 500MHz/2.5ns CCh.2 -500MHz/2.5ns CCh.3 1GHz/5ns CCh.4 -1GHz/5ns CCh.5 1GHz/10ns CCh.6 -1GHz/10ns

T Comment No.725 Comment It is unfortunate that we have an additional channel frequency table. Can this somehow be integrated into the existing frequency channel plan? Response Accept in principle. These are additional channels provided by the CoU pulses to support more SOP in addition to FDM. Remedy To distinguish from the representation of FDM channels, replace Ch.# by CCh.#. (Chirping Ch). (Table 39I)

T Comments No. 727 Comment Are of the parameters of table 39m represented in the MLME/PLME/MAC commands? Response Accept in principle. Remedy To be added. (See general framework)

T Comment No. 728 Comment Problem in notation between Equation 2 and Figure 27j. This may only be editorial ... is there an argument change between equation 2 and Figure 27j? Response Accept in principle. The argument is ‘tau’. See next page

T Comment No. 728 (Continue) Time [ns]

T Comment No.866 Comment Need details on how the transmitter and receiver will negotiate the use of the chirp pulses Response Accept in principle. Remedy This is controlled by PAN coordinator. See general framework.

T Comment No. 867 Comment For CS pulses, does the TX filtering affect the receiver performance? How does the receiver know which pulse shape to expect? Response Accept in principle. The receiver must support the same pulse as the transmitter. Remedy See general framework.

E Comment No.143 Comment Add "CoU", and all other new acronyms and abbreviations, to clause 4. Response Accepted Remedy Add as suggested.

E Comment No.255 Comment Be sure to add "SOP" to clause 4, Acronyms and Abbreviations. Response Accept Remedy Add as suggested.

E Comment No.272 Comment third paragraph; Which Equation (1)? Response Accept. Remedy Change to Equation (o-1). (Page 36, above the graph)

E Comment No.273 Comment SI unit for nanoseconds is ns Response Accept. Remedy Replace ‘nsec’ with ‘ns’. (Caption of Figure 27j)

E Comment No.274 Comment second paragraph; What is this jumble of words trying to state? Response Accept in principle Remedy Replace the paragraph above Table 39m with “Group delay values used for this option are listed in Table 39m. Each 500MHz band shall use No.1 or No.2 pulses, while Each 1.5GHz band shall use one of No.3 through No.6 pulses.” (Page 38)

E Comment No.440 Comment Equation (1) is the wrong cross reference. Response Accept. Remedy Replace ‘Equation (1)’ with ‘Equation (o-1)’

E Comments No.588 Comment 1. Replace 'Equation (1)' by 'Equation (o-1)'. 2. Replace 'chirped DS pulse' by 'CoU pulse'. Response Accept. Remedy Change as suggested. (above Figure 27h)

E Comments No.726 Comment I guess I don't understand the concept behind continuous spectrum pulses. Perhaps a little bit of explanation as to why these pulses are continuous. For example, what is a non-continuous spectrum pulse? Response Accept in principle. Remedy Change all occurrences of continuous spectrum pulses to controlled phase-continuous pulses in section 6.8a.3.3.5 and add the following text ahead of the paragraph above Equation (o-2). “This pulse shape is generated by passing a mandatory pulse through a filter with controlled group delay characteristics over the signal bandwidth. It is a group delay controlled pulse.”

Not Comment But … Remove the following text above Table 39I in 6.8a.3.3.4 “It should be noted that a 500MHz bandwidth implies the 494MHZ and 507MHz bandwidths and that a 1.5GHz bandwidth implies the 1.4GHz and 1.3GHz bandwidths, given in Section 6.1.2.1b.”