Microwave Link Design Trade-offs

Microwave Link Design Trade-offs: Throughput speed/frequency bandwidth/frequency band/latency/Ethernet frame size/TDM (STM1, E1 etc.)/Antenna heights/Tower heights/link distances/Diversity Loss

The following is a ‘Typical Requirement Scenario’ to introduce why trade-offs are needed during the link design process.

The requirement is often to achieve the highest possible throughput from a radio link.  Some practical implications of this requirement are as follows:

1)        Consider using XPIC (use vertical and horizontal polarisation with the same frequency channel to double the capacity of the channel)

2)         Consider using the larger bandwidths (either 28MHz or 56MHz)

3)         Higher-order modulation must be considered:

  1. The consequence: lower transmit power and higher receive sensitivity
  2. The consequence of a. is the need to use larger-sized antennas
  3. There is a possibility that the these large antennas cannot be accommodated on a building, tower or mast
  4. There is a possibility that links must be shorter in order to reduce required antenna sizes

4)        If the link distances are long, one needs to consider 6/7/8GHz frequencies.

5)        The trade-offs to be made are between:

Budget vs throughput vs frequencies vs channel bandwidth vs link distance vs antenna sizes vs protection (redundancy/diversity).

This process will entail a few iterations.  Various options must be compared against each other.

6)   Consider that for a voice-only network application, the throughput speed requirements will be less.  Smaller bandwidths such as 3.5MHz or 7MHz can be used.  The antenna sizes can be smaller and the achievable link distances can be longer.  It is better to match the radio backhaul network technology to the application requirement.  If the application requirement is voice-only, it may not make financial sense to design a high-bandwidth video-capable network.  Increased costs due to the need for large antenna and sturdier tower infrastructure may become prohibitive.

There are a number of questions to be addressed when considering deployment of Microwave or IP backhaul links.

  1. What service interfaces are needed – STM-1e, STM-1o, GbE, 10/100Base Ethernet, 10BaseT Ethernet, E1s (or T1s?).  Is 1E1, 2E1, 4E1, 8E1, 16E1 or 32E1 needed?
  2. For Ethernet:  is Synchronous Ethernet and compliance against IEEE1588v2 a requirement?  What are the QOS and OAM requirements?
    • What Ethernet frame sizes will traverse the link?  Throughput speed and latency vary with frame size.
    • Will 802.1Q VLAN tagging be used?  Will 802.1ad (informally known as QinQ) be used?
  3. What is the link’s throughput requirement in Mbps or Gbps?  You can determine this requirement from either the number of service interfaces or the end-application requirement.  For a dedicated radio network that carries voice traffic, the solution may only need the equivalent of a few Mbps – this will allow use of smaller bandwidths and therefore longer links, while minimizing the required antenna sizes.
  4. Are additional wayside services needed e.g. Network Management, RS232, V24, EoW (Engineering Order Wire), wayside E1s to complement transmission of STM-1(s) etc.
  5. What frequency band is to be used? 4/6/7/8/10/10.5/11/13/15/18/23/26/28/32/38/42GHz.  Note that there are other frequency band options that are not included in the above list of ITU-R specified bands.
  6. What T/R spacing is to be used.  Each band has a different number of T/R spacing options.  One must be selected:  please refer to the frequency table above.  There can be between one and five T/R spacing to choose from.
  7. What bandwidths are available in the chosen band?  In ETSI-regulated markets, microwave links typically work with 3.5MHz, 7MHz, 14MHz, 28MHz or 56MHz channels.
    • Now, given the throughput requirement (c), the frequency band (e) and bandwidth (g), using data throughput speed tables such as those in Appendix A, one can determine the modulation that needs to be used.  Since the modulation is known, one can determine the transmit power and receive sensitivity based upon radio specification tables.  Now, taking GPS location (hence link distance) into account, by implication, one can determine environmental effects on the microwave link. To ensure 99.999% (or 99.99x%) link availability, one now chooses suitable-sized antennas, while taking into account practicalities such as maximum size of the antenna that can be mounted on the building, mast or tower.
  8. Do you wish to use a:
    • ‘Split-mount System’ (where the Indoor Unit (IDU) is placed within an ‘equipment room’ or a container at ground level and the Outdoor Unit (ODU) is placed on the mast or tower adjacent to the antenna)
      • If so, what is the maximum distance between this IDU and ODU likely to be?  An IF cable will be run between the IDU and ODU.  This IF cable conveys DC power and the link data.
    • ‘All-Indoor System (AIS)’ (where all equipment is placed at ground level or at least ‘indoors’ in a temperature and humidity-controlled environment) where waveguide is used for transmission of microwave energy to/from the antenna (or antennas).
    • ‘All-Outdoor (AO) System’ (where baseband signals such as Ethernet connect directly to an Outdoor Unit (no IF frequencies originate from or are received within the ‘indoor environment’)
  9. What type of ‘link protection’ is required (if any) – the options include amongst others; 1+0 (no protection), 1+1 HSB, 1+1 SD, 1+1 FD, N+1, Ethernet Ring, E1 Ring, STM Ring.  Or, do you wish to use a 2+0 implementation which offers a multiplexing rather than diversity benefit to double a link’s throughput using 2 frequency channels?
  10. What is the link distance (also referred to as ‘range’)?
  11. What are the GPS coordinates of the sites for the links?
  12. Practical antenna-related constraints may impact throughput speed objectives.  Regarding the building, mast or tower, technical considerations include:
    • What are the building, mast or tower heights at each side of a link
    • Is spare space available at the ‘design height’ on the infrastructure for the new antennas?
    • What is the maximum antenna loading for all antennas on the mast or tower?  Will the addition of new antennas lead to an exceeding of the mast’s maximum permissible loading?  Will the addition of new antennas lead to excessive twisting and swaying of the antenna-support infrastructure?
    • Is there a constraint regarding the maximum antenna size allowed on the mast?
    • Does the building, mast or tower infrastructure allow addition of new supporting infrastructure (brackets, cabling etc.) for the new antennas?
  13. If we consider all of the above, antenna constraints permitting, can we achieve a 99.999% link availability design objective?  If antenna size and wind loading constraints are not a limitation and the antennas on each side of a link can be placed sufficiently high so that the link is only constrained by line-of-sight path losses and not the addition of excessive diffraction and multipath reflection losses, will the transmit power, receive sensitivity and antenna sizes be suitable to ensure 99.999% ‘link availability’ for the required throughput and link distance?  Each situation demands consideration of points (a) to (l) above.
    • If there is access to frequencies and bandwidth, one can trade-off between throughput, link distance and required antenna gain (hence antenna size)
    • For longer-distance links
      • one may need to sacrifice on the throughput objective in order to have practical-sized antennas (furthermore, too large antennas may not be permitted).
      • One may need to deploy 1+1 SD links (using this topology, two copies of the same data-stream are send simultaneously and the best-performing stream (least affected by multipath and fades) is chosen – this allows a maximising of ‘link availability’.
      • Specifically for 1+1 SD and long distance links, one must be aware that ‘horizon effects’ come into play – if masts or towers are not on mountain high-sites, the required antenna heights (hence mast or tower heights) can quickly exceed 100m.
    • A doubling of frequency or distance has a 6dB effect on the link’s free-space-loss budget.  Therefore, bandwidth availability permitting, for longer distance links one considers lower frequencies such as 6/7/8GHz.  An additional benefit of working at lower frequencies is that radio product maximum transmit power levels are higher.
    • Furthermore, in some cases, the use of XPIC technology (using vertical and horizontal polarisation on the same frequency channel) allows a doubling of a link’s throughput capacity.

The cost for the shipping of antennas is a critically important consideration.  Long-term planning is mandatory for microwave link projects.  If timescales are a project constraint, one must bear in mind that even for 0.6m High Performance antennas, the air freight shipping cost is more than the cost of the antenna itself.

Sagittar Microwave Radio Links – Africa, Middle East and SE Asia

Sagittar Microwave Radio

Sagittar Microwave Radio Link Solutions

Email us at [email protected]

Or, please follow this link to ask questions and request additional information.