TR-30.1/01-07-056 Telecommunications Industry Association (TIA) Baltimore, MD 16-20 July, 2001 COMMITTEE CONTRIBUTION Technical Committee TR-30 Meetings Source: Agere, AMD, Conexant, Broadcom, Intel Contributor: Name: Phone: E-mail: Project: PN-4643 Title: For Information: G.pnt.pml Contribution presented to May 2001 ITU-T Q4/15 Rapporteur Meeting Distribution: Members of TR-30.1 and meeting attendees Keith Chu +1 (949) 579-4121 keith.chu@mindspeed.com Abstract The attached contribution on G.pnt.pml was recently presented at the May 2001 Q4/15 meeting in Red Bank, NJ. It is presented here for information. The contributor grants a incorporate text contained publication; to copyright in contribution; and at TIA's standards publication. free, irrevocable license to the Telecommunications Industry Association (TIA) to in this contribution and any modifications thereof in the creation of a TIA standards TIA’s name any TIA standards publication even though it may include portions of this sole discretion to permit others to reproduce in whole or in part the resulting TIA Introduction As most members of TR-30.1 are no doubt aware, ITU-T Recommendation G.989.1, Phoneline networking transceivers – Foundation, was approved by SG15 earlier this year. TR-30.1 project PN-4643 was instrumental in progressing this work, and the source companies on this contribution wish to thank other TR-30 members for their assistance in this effort. G.989.1 defines the frequency spectrum, header and trailer format, and MAC procedures for PNT transceiver. It does not, however, define a payload format. An open project in Q4/15, G.pnt.pml, seeks to accomplish this. The attached contribution, defining modulation and constellation encoding for a G.pnt.pml layer, was presented at the ITU-T Q4/15 Rapporteur meeting in Red Bank, NJ. It was accepted by Q4 at that meeting as working text for inclusion in a proposed G.pnt.pml Recommendation. It is being presented at this meeting for information and in order to progress the work on PN-4643 and the G.pnt-series of Recommendations. The source companies on this contribution expect to present further contributions on G.pnt.pml at the August meeting of Q4/15 and the September meeting of TR-30, with the goal of reaching Consent on a G.pnt.pml Recommendation at the October meeting of ITU-T Study Group 15. 2 ITU - Telecommunication Standardization Sector Temporary Document RN-081R1 STUDY GROUP 15 Original: English Red Bank, NJ - 21-25 May 2001 Question: 4/15 SOURCE1: AMD Broadcom Conexant Systems ESS Technology Intel Lucent Technologies Motorola TITLE: G.pnt: Modulation for G.pnt.pml ________________________________________________________ ABSTRACT This contribution proposes text to be used in draft Recommendation G.pnt.pml. It describes the modulation to be used. 1 Introduction: The following text defines the modulation to be used in a G.pnt.pml frame; i.e., the portion of a G.989 frame bounded by the G.989.1 Frame Type octet, and the G.989.1 EOF octet. 2 Modulation The G.pnt.pml frame shall use Quadrature Amplitude Modulation (QAM). The two-dimensional symbol rate shall be 4,000,000 symbols/s, with a tolerance of ±0.01%. 4.2. Carrier Frequency and Tolerance The carrier frequency shall be 7 MHz. The carrier clock shall be locked to the symbol clock. 4.3. Transmit Filters Transmit filtering sufficient to meet the PSD mask defined in G.989.1 shall be applied. Note: Identical transmit filtering is used for all symbol mappings and bit encodings. 4.4. Transmitter Symbol Response The magnitude transmitter output shall be upper-bounded by the following temporal mask. The response shall be measured across a 100Ω load between tip and ring at the transmitter’s W1 interface. 1 3 Contact: Intel Corp. Barry O’Mahony T: +1 (503) 264-8579 E: barry.omahony@ties.itu.int Normalized Magnitude 1 0.4 0.25 0.07 0.056 0.015 0.01 0.005 0 0.045 0.1 0.15 0.23 0.8 1.4 2.5 4.0 5.0 t (sec) Figure 2.1Transmitter Symbol Response Magnitude Mask Output before t = 0 and after t = 5.0 microseconds shall be < 0.032% of the peak amplitude. 4.5. Symbol Mapping The G.pnt.pml frame shall use the 4D symbol mapping defined in § 4.5.5. Optionally, a portion of the G.pnt.pml frame may use the 2D symbol mapping defined in § 4.5.6. 4.5.5. 4D symbol mapping Each 4D symbol shall consist of two consecutive 2D symbols. The first 2D symbol shall be modulated with 2 to 8 data bits using the constellation encoding in § 4.6. The second 2D symbol shall be transmitted with zero amplitude. 4.5.6. 2D symbol mapping Each 2D symbol shall be modulated with 2 to 8 data bits using the constellation encoding in § 4.6. 4.5.7. 2D – to – 4D transition When switching from 2D to 4D symbol mapping within a given G.pnt.pml frame, an additional zero amplitude symbol shall be transmitted immediately following the last 2D symbol and immediately preceeding the first 4D symbol. 4.6. Constellation Encoder 4.6.5. Bit to Symbol Mapping The incoming bits shall be grouped into N-bit symbol intervals, where 2 ≤ N ≤ 8. Each N-bit group shall be mapped to a 2D symbol as shown in Figure 2.2 through Figure 2.8 The symbol values are shown with bits ordered such that the right-most bit is the first bit in time and the left-most bit is the last bit in time All constellations except for 3 bits-per-symbol lie on a uniform square grid, and all constellations are symmetric about the real and imaginary axes. The 3 bits-per-symbol constellation points are equally spaced around an equal-amplitude circle. Figure 2.2 – 2 bits-per-symbol 4 Figure 2.3 – 3 bits-per-symbol Figure 2.4 – 4 bits-per-symbol Figure 2.5 – 5 bits-per-symbol Figure 2.6 – 6 bits-per-symbol 5 Figure 2.7 – 7 bits-per-symbol 6 Figure 2.8 - 8 bits-per-symbol 4.6.6. Symbol Mapping Transitions Symbol mapping transitions within a G.pnt.pml frame shall occur on a data bit octet boundary, for transitions between 2D and 4D symbol mapping or for transitions in bits-per-symbol mapping. If the number of data bits mapped to the symbol immediately preceding such a transition is less than the current value of N, a sufficient number of bits of value zero shall be appended to the data bits in order to complete this last symbol in the sequence. 4.6.7. Scaling The relative scaling of different constellations is given by Table 1. The constellation points shall be scaled such that the reference points have the values shown, within a minimum-distance tolerance of ±4%. 7 Bits per Symbol 2 Reference Point 00 4D mapping amplitude 1 3 000 1 4 0000 5 00000 6 000000 7 0000000 8 00000000 2D mapping amplitude 1 3 1 4 1 7 1 9 1 15 1 2 1 2 1 3 2 1 4 2 1 7 2 1 9 2 1 15 2 Table 1 Constellation Reference Points Note: The constellation points are scaled such that the outermost points have approximately equal magnitude. 3 Summary: This paper should be presented under the G.pnt agenda item. It is proposed that the above text be used in draft Recommendation G.pnt.pml as a description of the modulation method. 8