No. 1 "低空掠海飞行能力差，超音速导弹由于音爆反射问题..." It is true that hypersonic is not very good at flying low, but not because of sonic boom reflection. Sonic boom is a weak shockwave, and the reflection is even weaker. As long as the vehicle's velocity relative to the ground is greater than the critial Mach, the reflection can't catch the vehicle. The reflection only hits the vehicle when it turns or something. The sonic boom enercy on the ground is pretty hard to compute. Experiments are done on measure sonic boom generated by F-4 flying around 100ft is about 21 psf (pound per sqrt foot, many aerodynamic measurement in the US are still in imperial standard, which I hate, but I have to deal with them everyday). The reflection of it is of course smaller, and the measurement has done for that. The problem of flying hypersonic relatively to ground in low altitude is that the time-to-die (when flying level fly, the time to take to impact the ground is time-to-die) is a fraction of a second, there is no room for error. A little bumpy air will cause the vehicle to crash. Hence hypersonic and low altitude, like lower than 500ft is not realistic. No. 4, is definitely correct. One thing to add, even for a boost guide vehicle, which isn't powered, still the infra-red signature is like a big bullseye. No. 5, GNC (guidance, navigation and control) for subsonic cruise missile is harder than those in supersonic or hyper sonic. Since errors are accumulative, longer the flight, errors will occure. Since all measurements have errors, they show up and impacting calculations. Hence GNC for subsonic cruise missile does a lot more check, and the code is a lot more difficult. It is very very hard to intercept, say a tomahawk. You cannot intercept something you cannot see, no matter how slow it is. Tomahawk and those subsonic cruise missile flight below the radar. Even someone tells you with a cell phone that a cruise missile is coming, you won't be able to intercept it. By the time you see it with your eyes, there isn't anything you can do.
Orbital manuever olny happens in space, where there is no atmophere. The only way it can change its orbit is by the thrusters. On the other hand, when after reentry, those glide vehicles like DF-21, I assume, are lifing bodies, and they glide through air and perform manuever, and they don't follow orbits. How can they change something which isn't there in the first place? Glide vehicles can change its trajectories, and that is what you want to say.
且不说下面视频美军的GQM-163土狼靶弹是业界公认的全球性能最好的超音速反舰导弹，飞的更低更远，能模拟各种刁钻机动，性能完虐中俄超音速导弹，单说美军60年代竞标失败的沃特公司的ALVRJ超音速导弹，技术转让台湾后有了雄风3，射后不理，一百公里外自主寻找目标击中几十吨重渔船。军迷只顾调侃，军委却是直冒冷汗：若在大军横渡台湾时，台湾向台海盲射数百枚雄风3，自主选目标攻击，后果不敢想象的。恐怕连小小的登陆艇都无法幸免的。
业界素有超音速与亚音速反舰之争。美国经过长期探索，最终选择了亚音速反舰导弹作为主要技术手段，原因只有一个：超音速反舰导弹的突防能力远不如亚音速导弹，作战效果不佳。速度再快，被拦截了，岂不是白费蜡？不过美国目前依然装备着大量的超音速反舰导弹，当遇到对抗能力不强的对手时就会使用。美军亚音速反舰导弹都是专用反舰导弹，对付强敌的。而超音速的则是防空反舰两用导弹，标2，标3，标6都是两用弹的。
美国自50年代发展的黄铜骑士，鞑靼人，波马克，标准都是超音速反舰导弹。波马克采用高空弹道时速度高达4马赫，射程高达1000公里，低空弹道600公里，速度3.5马赫。但波马克都能击落波马克，标准也可以拦截标准，没有什么特别的突防威力，只要是迎击状态，目标速度根本不需要考虑，只要雷达能稳定跟踪就行。按波马克的性能来看，中俄在冲压发动机方面的技术还差的远，连美国50年代末的水平都没有达到的。
超音速反舰导弹的优点就是飞行速度快，理论上动用超音速反舰导弹攻击目标，留给对方的反应时间更短，但实际应用上根本不是这样一回事。技术障碍主要有以下几点；
1. 低空掠海飞行能力差，超音速导弹由于音爆反射问题，导弹体积越大就越严重，通常5米长度的导弹降低到15米以下高度持续平飞的话，弹体就会产生震动，时间稍长就会导致导弹失控坠海或引导头失效。而亚音速导弹则可轻松在1米左右波峰高度长时间飞行。目前防空导弹的近炸引信在2米以下高度根本无法区别小目标与大海杂波。就我所知，只有标准六很好解决了此问题，采用一种很巧妙的方式，这是后话。
2. 多谱勒效应问题，由于速度快多谱勒效应明显，雷达可以不费吹灰之力从海面杂波中捕获到导弹的回波，而且信号清晰跟踪稳定，即使掠海飞行都会被发现。而亚音速导弹多谱勒效应不明显，对于雷达来说很难从海面杂波中截获到，即使能够发现导弹也很难稳定的跟踪，信号时断时续，很难远距离的拦截。
3. 无法采取隐身设计，因为导弹要高速飞行，弹体根本就不可能采用棱锥型或截锥型隐身设计。即使退一万步，找到弹体隐身方式，但超音速目标裹的一层压缩空气仍然能被精密雷达探测出来，那又有什么用？F22只要进入超音速就隐身性大降，就是因为这回事，F22进入与离开战场是超音速，但进入战场后就要转为亚音速，发挥他的隐身能力
4. 热效应问题，超音速弹必须有强大的推力，目前的技术就是使用火箭或冲压发动机等推力越大发热越明显的发动机，尾焰又长又明显，加上空气摩擦的热效应，根本不可能不被光学探测仪器红外探测器发现。
5. 引导系统可靠性问题，解释起来复杂，简单一句话概况，因飞行条件的不同，亚音速导弹在控制精度，捕获目标能力以及遭到电子干扰时的反干扰能力都要超过超音速导弹。
6. 末段机动能力有限，躲避拦截的最有效手段就是末段机动，超音速导弹速度太快，过载太大，机动幅度一大，很可能就错过目标了，而亚音速导弹末段机动可以设定的非常复杂刁钻，错过目标后可以重新返回接战（所谓的回马枪）。爱国军盲常吹嘘的东风末端10马赫高机动更是天方夜谭，不怕闪了腰？想机动就得降速，只能二选一。
7. 弹体加强的潜力，亚音速导弹可以安装装甲，遭到拦截后如果受损不大的话，仍然可以击中目标，但超音速导弹加固到与亚音速导弹同样的抗打击能力则显然不可能，披挂上一层厚装甲的话，在动力方面付出的代价要高的多。此外遭到同样的拦截弹命中，亚音速的导弹可以当是飞机停在地面给个鸡蛋砸一下，而超音速的导弹则要被洞穿解体，再厚装甲也木有用的。
目前舰艇的雷达与光学探测器都能在50公里甚至更远距离发现并锁定超音速导弹。如果采用高空弹道，在200公里的距离上就能捕获信号。在白天用肉眼也能在7-8公里距离发现导弹尾焰，夜间则20公里距离就能用肉眼发现导弹尾焰 。毛子国用SA-N-4舰载防空导弹测试，结果拦截SS-N-22超音速导弹试验做到4发4中，但对SS-N-25亚音速导弹则完全没有拦截能力，可见超音速反舰导弹不但留给对方的反应时间比亚音速导弹还长的多，而且拦截起来更要容易的多，与理论完全满拧。
争议归争议，让我们看看实战效果：
亚音速反舰导弹 - 马岛战争时英军谢菲尔德号的雷达发现了掠海来袭的飞鱼导弹，但信号时断时续根本无法锁定，8秒后中弹。第4次中东战争中以色列对埃及与叙利亚海军舰艇发射加伯列导弹，没有1艘舰艇的雷达发现了导弹，全部都是了望手肉眼发现导弹，2-3秒后中弹。还有鱼叉击沉的多艘军舰，此类例子比比皆是。
超音速反舰导弹 - 唯一的战例还是客串的，美军的护卫舰对伊朗乔森号导弹快艇发射了一枚标准2导弹，后者虽然用雷达发现了导弹，但因没有拦截与干扰能力，只能进行急剧机动，然尔无法摆脱，结果被击沉。其它超音速弹的战绩至今为零
美海军舰艇官兵曾跟我交心：
拦截训练时，额们不怕超音速导弹，就怕这个慢吞吞贴着浪尖飞，油盐不进的亚音速瘪犊子。飞的低，防护好，静悄悄的，跟鬼子进村一样，令人头疼
额们最喜欢中俄的带冲压发动机的超音速导弹，动静大，飞不低，那热量跟灯泡一样，都不用雷达测，用红外探测器即可，离着大老远就像打着手电，敲锣打鼓而来，简直就是标准6的盘中餐
最后聊一句标准六，它的战斗部是模块化的，防空时用60公斤MK125破片弹头，反舰时则用上百公斤的串联式聚能破甲弹头，与你们想象不一样的。东风在标准6面前高机动有用么？看看人家的打靶就知道了(4:09)：来袭导弹做出各种刁钻末端机动，还是被标6一箭穿心
就像红旗7，既能在海上拦截反舰弹，也能在陆地上拦截陆攻巡航导弹。
No. 1 "低空掠海飞行能力差，超音速导弹由于音爆反射问题..." It is true that hypersonic is not very good at flying low, but not because of sonic boom reflection. Sonic boom is a weak shockwave, and the reflection is even weaker. As long as the vehicle's velocity relative to the ground is greater than the critial Mach, the reflection can't catch the vehicle. The reflection only hits the vehicle when it turns or something. The sonic boom enercy on the ground is pretty hard to compute. Experiments are done on measure sonic boom generated by F-4 flying around 100ft is about 21 psf (pound per sqrt foot, many aerodynamic measurement in the US are still in imperial standard, which I hate, but I have to deal with them everyday). The reflection of it is of course smaller, and the measurement has done for that. The problem of flying hypersonic relatively to ground in low altitude is that the time-to-die (when flying level fly, the time to take to impact the ground is time-to-die) is a fraction of a second, there is no room for error. A little bumpy air will cause the vehicle to crash. Hence hypersonic and low altitude, like lower than 500ft is not realistic. No. 4, is definitely correct. One thing to add, even for a boost guide vehicle, which isn't powered, still the infra-red signature is like a big bullseye. No. 5, GNC (guidance, navigation and control) for subsonic cruise missile is harder than those in supersonic or hyper sonic. Since errors are accumulative, longer the flight, errors will occure. Since all measurements have errors, they show up and impacting calculations. Hence GNC for subsonic cruise missile does a lot more check, and the code is a lot more difficult. It is very very hard to intercept, say a tomahawk. You cannot intercept something you cannot see, no matter how slow it is. Tomahawk and those subsonic cruise missile flight below the radar. Even someone tells you with a cell phone that a cruise missile is coming, you won't be able to intercept it. By the time you see it with your eyes, there isn't anything you can do.
Orbital manuever olny happens in space, where there is no atmophere. The only way it can change its orbit is by the thrusters. On the other hand, when after reentry, those glide vehicles like DF-21, I assume, are lifing bodies, and they glide through air and perform manuever, and they don't follow orbits. How can they change something which isn't there in the first place? Glide vehicles can change its trajectories, and that is what you want to say.